Background Carbapenems (CBPs) are being used more and more because of the increasing prevalence of extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae. Due to the extensive misuse of these antibiotics, some bacteria have developed CBP-resistant mutations. This epidemiological situation should make us wonder about prescribing CBPs.

Purpose To describe prescribing patterns of imipenem/cilastatin, ertapenem and meropenem in elderly inpatients: context and impact of an interdisciplinary approach to prescriptions analysis.

Materials and Methods A retrospective study of CBP prescriptions was performed over a ten-month period (March-December 2011) in geriatric departments (313 beds). Data were collected from the electronic medical records, bacteriological analysis results and email exchanges between the infectious diseases physician (IDP), bacteriologists and pharmacists (prescription monitoring system). The following items were noted: patients, prescriptions and bacteriological characteristics.

Results 55 patients were included with a total of 61 CBP prescriptions. The mean age was 83 (sex ratio 0.72). 71% of patients accumulated between 2 and 5 risk factors of multidrug resistant bacteria. Imipenem was the most-used carbapenem (n = 35; 57%) compared to ertapenem (n = 23; 38%) and meropenem (n = 3; 5%). Major indications were urinary tract infections (n = 37; 61%) and pneumonia (n = 15; 25%). 59% of infections were nosocomial. 39% of CBP prescriptions were written after a first-line antibiotic had failed (ceftriaxone most of the time). The overall duration of carbapenem therapy was 11 days. Microbiologically-documented infections and ESBL bacteria accounted for 69% (n = 42) and 51% (n = 24) of prescriptions, respectively: 5 of the ESBL strains isolated were community-acquired bacteria. 61% (n = 38) of prescriptions were reassessed by an IDP: 29 (76%) were in accordance with recommendations; 7 (18%) were stopped or changed for a narrow-spectrum antibiotic.

Conclusions CBP prescriptions seem relatively well controlled in geriatric care units due to multidisciplinary analysis of the prescriptions. Nevertheless, evaluation of the impact of monitoring prescriptions for use of CBPs requires longer follow-up.

No conflict of interest.